Joint spectrum sensing and resource allocation optimization using genetic algorithm for frequency hopping–based cognitive radio networks

Abstract

In cognitive radio (CR) networks, secondary users should effectively use unused licensed spectrums, unless they cause any harmful interference to the primary users. Therefore, spectrum sensing and channel resource allocation are the 2 main functionalities of CR networks, which play important roles in the performance of a CR system. To maximize the CR system utility, we propose a joint out-of-band spectrum sensing and operating channel allocation scheme based on genetic algorithm for frequency hopping-based CR networks.In this paper, to effectively sense the primary signal on hopping channels at each hopping slot time, a set of member nodes sense the next hopping channel, which is called out-of-band sensing. To achieve collision-free cooperative sensing reporting, the next channel detection notification mechanism is presented.Using genetic algorithm, the optimum sensing and data transmission schedules are derived. It selects a sensing node set that participate the spectrum sensing for the next expected hopping channel during the current channel hopping time and another set of nodes that take opportunity for transmitting data on the current hopping channel. The optimum channel allocation is performed in accordance with each node's individual traffic demand. Simulation results show that the proposed scheme can achieve reliable spectrum sensing and efficient channel allocation.

Guided joint spectrum sensing and resource allocation using a novel random walk grey wolf optimization for frequency hopping cognitive radio networks

“…The sensing overload is reduced by selecting specific member nodes by CH to perform sensing. Out‐of‐band sensing is demonstrated in Figure . Black big rectangles represent primary signals and the small packets are shown with small rectangles, which represent the SUs.…”

“…Without disturbing primary signals and effectively using the spectrum holes, accurately sensing the spectrum is one of the major function of CRNs. Spectrum sensing can be performed by any spectrum‐sensing mechanism like energy detector, cyclostationary feature detector, and matched filter . The secondary signals should take care of primary detection probability ( P d ) and false alarm probability( P fa ) during the time of spectrum sensing.…”

“…Energy detector is assumed to be used in this scenario. The probability of detection and probability of false alarm of node for each and every narrow‐band channel can be mathematically shown as in Equation : where T = and γ is a threshold. While setting the false alarm for the first time, the threshold γ is determined by Equation .…”

“…While setting the false alarm for the first time, the threshold γ is determined by Equation . Minimum number of sensing samples N min can be determined from Equation as a function of SNR, where SNR = …”

“…Whereas, in the proposed algorithm, the nodes sensed only the specific channel, hence, T r is equal to the time duration of PUDN signal. The service disruption time ( T SD ) is given as in Equation : …”